Introduce new generic ADC driver for ADS1115 chip

[SalimTerryLi]

Describe problem solved by this pull request
This driver would provide ADC interface for boards without on-chip ADC function such as RPi.

Describe your solution
Sampling on trigger, SPS=250, VREF=6.144.
Scheduled at 80Hz, looping through its four multiplexed channel, the driver publishes a new adc_report after all 4 channels are captured, at 20Hz.

This driver is designed to run on any platforms. But ADC subscribers are not checking devID at now, so it should only run on boards without on-chip ADC.

Test data / coverage
Works fine on RPi.

[dagar]

I don't quite understand the need for the I2C_Interface and the ADS1115_Wrapper. Why not just have a simple ADS1115 that extends device::I2C?

Example: https://github.com/PX4/Firmware/blob/master/src/drivers/magnetometer/isentek/ist8308/IST8308.hpp

The other minor thing is the placement within another module (drivers/adc). I'd like to avoid modules within modules, although obviously our organization system needs work (where to put or name the generic adc driver).

[SalimTerryLi]

I don't quite understand the need for the I2C_Interface and the ADS1115_Wrapper. Why not just have a simple ADS1115 that extends device::I2C?

Example: https://github.com/PX4/Firmware/blob/master/src/drivers/magnetometer/isentek/ist8308/IST8308.hpp

The other minor thing is the placement within another module (drivers/adc). I'd like to avoid modules within modules, although obviously our organization system needs work (where to put or name the generic adc driver).

I just want to make it more clear to understand. ADS1115 is only a logic driver and the wrapper do the other things like device registering and scheduling. PCA9685(#14269) is also designed this way.
At least I want to do the PX4 specified scheduling out of driver logic.

Seems it cannot be achieved now.... I'll migrate them into the same place

[stale]

This issue has been automatically marked as stale because it has not had recent activity. Thank you for your contributions.

[SalimTerryLi]

updated & checked for availability

[dagar]

Happy to merge once #14477 is in.

[SalimTerryLi]

Happy to merge once #14477 is in.

Ready.

[dagar]

This is nitpicky, but for consistency how about keeping the folder name lowercase like all the other drivers and modules?

[SalimTerryLi]

Confirmed.

[ibrhm0v]

Everything seems ok to me @SalimTerryLi great work.
Update: The only issue I have faced testing your driver is when the driver is already running and is started again, it crashes and the driver can never be started again unless the autopilot is rebooted.

[SalimTerryLi]

Everything seems ok to me @SalimTerryLi great work. The only issue I have faced testing your driver is when the driver is already running and is started again, it crashes and the driver can never be started again unless the autopilot is rebooted.

Oh yes. It seems I forget to do ScheduleClear() on exit. I will test and fix it later. Thanks for your report.
Update: This comment is wrong.

[SalimTerryLi]

Everything seems ok to me @SalimTerryLi great work. The only issue I have faced testing your driver is when the driver is already running and is started again, it crashes and the driver can never be started again unless the autopilot is rebooted.

Could you please provide more information? How it crashed? Anything outputted? It works just fine on my Pi. BTW there is no need to manually call ScheduleClear();, as it is done by I2CSPIDriverBase.

pxh> ads1115 start -I
WARN  [SPI_I2C] Already running on bus 1
WARN  [SPI_I2C] UnknownApp: no instance started (no device on bus?)
Command 'ads1115' failed, returned -1.
pxh> ads1115 start -I
WARN  [SPI_I2C] Already running on bus 1
WARN  [SPI_I2C] UnknownApp: no instance started (no device on bus?)
Command 'ads1115' failed, returned -1.
pxh> ads1115 INFO  [ecl/EKF] 624834243: EKF aligned, (baro hgt, IMU buf: 18, OBS buf: 14)
pxh> ads1115 stop
INFO  [ads1115] stopping
pxh> ads1115 stop
ERROR [SPI_I2C] Not running
Command 'ads1115' failed, returned -1.
pxh> ads1115 start -I
ads1115 #0 on I2C bus 1
pxh> ads1115 start -I
WARN  [SPI_I2C] Already running on bus 1
WARN  [SPI_I2C] UnknownApp: no instance started (no device on bus?)
Command 'ads1115' failed, returned -1.
pxh> ads1115 stop
INFO  [ads1115] stopping

[ibrhm0v]

Oh ok it is my bad. I am working on my customized firmware that is still back at 1.9.0. Therefore I have ammended your main function. My start() implementation was wrong. I have fixed it now.